Such a detector is known from the article “Optical enhancement of diode laser-photo acoustic trace gas detection by means of external Fabry-Perot cavity” by Rossi et al., published in Applied Physics Letters, vol. 87, 2005. The detector described therein sends a chopped laser beam through a gas contained in an acoustic cell. The laser wavelength is tuned to excite particular molecules of the gas into a higher energy level. This excitation leads to an increase of the thermal energy, resulting in a local rise of the temperature and the pressure inside the acoustic cell. If the chopping frequency matches a resonance frequency of the acoustic cell, the pressure variations result in a standing acoustic wave. The acoustic waves are detected by a microphone in the acoustic cell. The resonance frequency of such an acoustic cell is typically between 500 Hz and a few kHz. In the detector of Rossi et al., a chopping frequency of 2.6 kHz is used for matching the resonance frequency of the acoustic cell.
Rossi et al. also describe using a Fabry-Perot cavity for amplifying the light intensity in the acoustic cell. The amplification is very advantageous because the sensitivity of the detector is proportional to the laser power. A feedback signal is obtained from a photodiode placed behind the Fabry-Perot cavity. In order to produce this feedback signal, the laser wavelength is weakly modulated by adding a small sinusoidal waveform to the power supply current. When the feedback signal exceeds a fixed threshold, the wavelength of the laser beam or the length of the cavity is adapted to bring the transmission back to the desired level. The photodiode does not provide a feedback signal when the chopper obstructs the passage of the light beam. As a result, the system can run out of its lock-loop for sudden disturbances, because it cannot respond sufficiently fast. In order to prevent this, a low response time is necessary such that the effect of harmonically turning the feedback signal on and off will be averaged out.
An important application of photo acoustic trace gas detectors is breath testing. Breath testing is a promising area of medical technology. Breath tests are non-invasive, user friendly and low cost. Prime examples of breath testing are monitoring of asthma, alcohol breath testing and detection of stomach disorders and acute organ rejection. First clinical trials show possible applications in the pre-screening of breast and lung cancer. These volatile biomarkers have typical concentrations in the parts per billion (ppb) range. Nitric oxide (NO) is one of the most important trace gases in the human breath, and elevated concentrations of NO can be found in asthmatic patients. Currently, exhaled NO levels at ppb concentrations can be only measured using expensive and bulky equipment based on chemiluminescence or far infrared optical absorption spectroscopy. A compact, hand-held, and low-cost NO sensor forms an interesting device that can be used to diagnose and monitor airway inflammation and can be used at the doctor's office and for medication control at home.
It is the challenge for these hand-held gas-analyzing devices to combine sufficient high sensitivity (ppb level) with a high robustness. Current photo acoustic trace gas detectors have the disadvantage that small form factor lasers (i.e. diode lasers) do not have sufficient laser power to reach the sensitivity required for trace gas detection. The use of an optical power enhancement cavity as described by Rossi et al. could increase the optical power, but that would lead to a slow feedback. When an optical enhancement cavity is used in combination with photo acoustics, the feedback signal is turned on and off intermittently by the chopper. Consequently, the cavity locking mechanism becomes very slow, which results in a system that is not robust enough for portable applications.